Hepatic capillariasis in rats:
a new model for testing
antifibrotic drugs
1Centro de Pesquisas Gonçalo Moniz, Fundação Oswaldo Cruz, and 2Faculdade de Medicina, Universidade Federal da Bahia, Salvador, BA, Brasil
M.M. de Souza1, L.M. Silva1,
A.A. Barbosa Jr.1,
I.R. de Oliveira2, R. Paraná2
and Z.A. Andrade1
Abstract
Rats infected with the helminth Capillaria hepatica regularly develop septal hepatic fibrosis that may progress to cirrhosis in a relatively short time. Because of such characteristics, this experimental model was selected for testing drugs exhibiting antifibrosis potential, such as pentoxifylline, gadolinium chloride and vitamin A. Hepatic fibrosis was qualitatively and quantitatively evaluated in liver samples ob-tained by partial hepatectomy and at autopsy. The material was submitted to histological, biochemical and morphometric methods. A statistically significant reduction of fibrosis was obtained with pentoxifylline when administered intraperitoneally rather than intra-venously. Gadolinium chloride showed moderate activity when ad-ministered prophylactically (before fibrosis had started), but showed a poor effect when fibrosis was well advanced. No modification of fibrosis was seen after vitamin A administration. Hydroxyproline content was correlated with morphometric measurements. The model appears to be adequate, since few animals die of the infection, fibrosis develops regularly in all animals, and the effects of different antifibrotic drugs and administration protocols can be easily detected.
Correspondence
Z.A. Andrade
Centro de Pesquisas Gonçalo Moniz FIOCRUZ
Rua Valdemar Falcão, 121 40295-001 Salvador, BA Brasil Fax: +55-71-356-4292 E-mail: zilton@cpqgm.fiocruz.br Received April 5, 2000 Accepted August 2, 2000 Key words ·Capillaria hepatica ·Hepatic fibrosis ·Septal fibrosis ·Antifibrotic drugs Introduction
Several animal models of hepatic fibrosis have been utilized for testing drugs exhibit-ing antifibrotic potential. The models of he-patic fibrosis utilized are mainly those ob-tained in rats by repeated injections of car-bon tetrachloride, by injections of porcine serum or following total main bile duct ob-struction (1). Mice infected with Schistoso-ma Schistoso-mansoni have also been used (2). How-ever, all of these models present limitations, especially due to individual variation in
fi-brosis development and high mortality, not counting the prolonged time involved and the high cost of some of them. Recently, it has been shown that rats chronically infected with the helminth Capillaria hepatica regu-larly develop progressive and diffuse septal hepatic fibrosis (3) similar to that seen in rats repeatedly injected with whole porcine se-rum (4,5). Fibrosis starts around the 40th day after infection, when the worms are already dead and the parasite-dependent, focal ne-crotic and inflammatory lesions are undergo-ing fibrous encapsulation and resorption.
Because septal fibrosis develops within a relatively short time (40-45 days) in 100% of the infected rats and shows a progressive course toward cirrhosis, this model has been considered to be one of the most adequate for testing antifibrotic drugs both in terms of their therapeutic and prophylactic effects.
This report concerns the response of he-patic fibrosis associated with C. hehe-patica infection to several drugs exhibiting antifi-brotic potential. It was assumed that the results obtained would be worthwhile both for testing the model and the drugs.
Material and Methods
Wistar rats of both sexes weighing 170 to 300 g were infected with approximately 1,000 embryonated eggs of C. hepatica each, ad-ministered by gavage. The eggs were ob-tained from the livers of experimentally in-fected rats. The liver tissue was homog-enized in a blender at 1,000 rpm, washed repeatedly in tap water, and left to sediment, until the supernatant fluid was completely clear. The sediment containing the immature eggs was placed on a Petri dish and covered with gauze humidified with 0.5% formalin. After 28-30 days at room temperature (26-28oC), the eggs became embryonated and
were recovered and used to infect the ani-mals. After inoculation the animals were divided at random into 5 experimental groups of five animals each as follows:
Group I. Infected control. The animals
received two weekly subcutaneous (sc) in-jections of 0.85% saline.
Groups II and III. The animals were treated daily with 6 mg of pentoxifylline (Trental; Hoechst, São Paulo, SP, Brazil), administered intraperitoneally (ip) and in-travenously (iv), respectively.
Group IV. The animals were treated with 1 ml of a 4 mM solution of gadolinium chloride (Sigma-66H3405; Sigma Chemical Co., St. Louis, MO, USA) administered iv.
Group V. The animals were treated with vitamin A (Arovit; Roche, São Paulo, SP, Brazil) twice a week by the sc route at the dose of 50,000 IU up to a total of 200,000 IU. Treatment was started for all the groups on the 25th day after inoculation. To follow the course of treatment and to evaluate the preventive potential of the drugs, the ani-mals were submitted to partial hepatectomy (surgical liver biopsy) one week after the end of treatment (45th day of infection). Treatment was resumed one week later, up to the time of sacrifice on the 95th day following infection. Table 1 summarizes the experimental protocols used.
For histological study, fragments of the liver collected during all the experimental phases were immediately fixed in Bouins fixative for 6 h, preserved in 70% alcohol and routinely embedded in paraffin. Sections were stained with hematoxylin and eosin. In addi-tion, the sirius red method for collagen was also used (6). The slides were blindly and independently evaluated by two pathologists
Table 1 - Schedule of treatment of fibrosis in rats with septal hepatic fibrosis associated with Capillaria
hepatica infection.
*Time considered from the day of inoculation.
Group Dose Route Duration of Hepatectomy* Sacrifice*
treatment (days)* I - Saline 1 ml sc 25-45 45th and 65th 95th II - Pentoxifylline 6 mg ip 25-45 45th 95th III - Pentoxifylline 6 mg iv 25-45 45th 95th IV - Gadolinium chloride 4 mM iv 25-65 45th 95th V - Vitamin A 50,000 IU sc 25-39 44th 108th
by a semi-quantitative method, which consid-ered fibrosis as absent (0), mild (+), moderate (++) and marked (+++) (see Figure 1). In case of disagreement, a consensus was reached before decoding the slides. Other portions of the liver were separated and submitted to the method of Bergman and Loxley (7) for meas-urement of hydroxyproline content.
Morphometric measurements were made on histological slides stained by the sirius red method and submitted to a computer image analyzing system (Leica Quantimet Q500MC, Cambridge, England). A total sec-tional area of 17.01 x 106 µm2 per case was
evaluated. A Leica Microstar IV microscope with a 4X objective was used to examine 9 microscopic fields at random. The sectional area of the fibrous tissue, red stained, was directly measured and calculated as percent of the total area examined.
An extra group was used to test the effect of gadolinium chloride on Kupffer cells. Normal rats were injected with a single dose or with 3 repeated doses of gadolinium chlo-ride (4 mM/animal) one every 72 h. Twenty-four hours after the last injection, the ani-mals received 1 ml of 10% India ink (colloi-dal carbon) iv.
Normal rats injected with India ink served as controls. The animals were sacrificed 24-48 h after injection of India ink and their livers submitted to histological examination by means of paraffin sections lightly stained with hematoxylin.
Results
Scattered focal lesions containing disin-tegrating parasites and their eggs varied in quantity but septal fibrosis was diffuse and appeared regularly in all infected untreated animals. By the 107th day of infection, the focal lesions had become small, encapsu-lated and frequently calcified, while septal fibrosis was more marked, being sometimes associated with areas of nodular regenera-tion of the hepatic parenchyma (Figure 1A).
Evaluation of the extent of septal fibrosis by a semi-quantitative histological method demonstrated that the groups treated ip with pentoxifylline exhibited a considerable de-crease or even total absence of septal fibro-sis (Figure 1B, C and D) as compared to the
Figure 1 - A, Disintegrating adult worms (Capillaria hepatica) are seen at the center of the picture, surrounded by a necrotic-inflammatory reaction and a fibrous capsule. Septal fibrosis and regenerating hepatic nodules are also present. H & E, 120X (bar represents 100 µm). B, Representative of the mild (+) degree of septal fibrosis observed in rats treated with pentoxifylline by ip administration. Septa are few, thin and incomplete, and the majority terminate abruptly within the hepatic parenchyma. Sirius red staining, 100X (bar represents 100 µm). C and D, Representatives of the moderate (++) and marked (+++) degrees of septal fibrosis observed in infected rats treated with antifibrotic drugs. C, Septa circum-scribe irregular areas of the hepatic parenchyma and show points of increased thickness (treated with gadolinium chloride); D, marked septal fibrosis delimiting hepatic nodules (cirrhosis) and forming areas of condensation, treated with vitamin A. Sirius red staining, 100X (bars represent 100 µm).
control group which presented a marked de-gree of fibrosis.
Morphometric evaluation confirmed the results obtained by semi-quantitative histol-ogy. The mean values for representative sec-tional areas of the liver revealed a consider-able reduction of fibrosis for the groups treated with pentoxifylline (given ip), in com-parison to untreated controls (P<0.005) (Table 2). The administration of vitamin A failed to modify the degree of liver fibrosis in a significant manner.
Hydroxyproline concentration (Table 3) also agreed with the data mentioned above but, in addition to indicating a statistically significant decrease in fibrosis for the group treated with pentoxifylline, it also showed similar results for gadolinium when admin-istered before fibrosis had become estab-lished (prophylactic treatment).
The uptake of colloidal carbon by sinu-soidal cells in normal rats treated with gado-linium chloride failed to show any
differ-ence when compared to untreated controls, with a similar number of sinusoidal cells containing carbon particles observed in both cases.
Data were analyzed by one-way ANOVA and the Student-Newman-Keuls test for group comparison.
Discussion
The model of septal fibrosis developed in rats infected with C. hepatica appeared to be adequate for testing antifibrotic drugs. He-patic fibrosis was constant, uniform and pro-gressive in infected controls and was signif-icantly, albeit variably, affected by antifi-brotic treatment. Since septal fibrosis does not appear before the 30th day of infection, drugs can be reliably tested with the present model for their prophylactic value when ad-ministered around the 25th day of infection or for their therapeutic action when adminis-tered after the 50th day.
Clear-cut positive results were obtained with pentoxifylline treatment, with the route of drug administration appearing to be cru-cial. The ip route yielded better results than the iv route. It is rather difficult to explain this finding, since the action of pentoxifylline seems to be quite complex. Pentoxifylline (methylxanthine) is known to inhibit col-lagen synthesis, suppress Kupffer cell acti-vation and promote decrease of serum TNF-a (8,9). It cTNF-an TNF-also inhibit collTNF-agen deposi-tion and connective tissue cell proliferadeposi-tion (10). Probably, the plasma concentration and half-life of the drug differ according to the route of administration. Also, when injected ip, the drug would reach sinusoidal cells in the liver more rapidly or at more adequate concentrations than when systemically ad-ministered.
Sinusoidal cells play a pivotal role in septal fibrosis (11). As a matter of fact, the possibility of the drug interfering with sinu-soidal cells stimulated us to use gadolinium chloride in this investigation. Gadolinium is
Table 3 - Comparison of the mean values obtained by measurement of hepatic hydroxyproline content in the various experimental groups.
*P<0.05 compared to control (Group I) (Student-Newman-Keuls’ test). N = 5 for all groups.
Groups Mean ± SD (µm2) Treatment
I 8.84 ± 1.91 PBS (0.85%)
II 5.69 ± 0.49* Pentoxifylline (ip)
III 7.85 ± 1.54 Pentoxifylline (iv)
IV 6.41 ± 0.89* Gadolinium chloride (preventive)
V 7.08 ± 0.97 Vitamin A
Table 2 - Comparison of the mean values obtained by morphometric measurement of fibrosis in liver sections stained with sirius red from the various experimental groups. *P<0.05 compared to control (Group I) (Student-Newman-Keuls’ test). N = 5 for all groups.
Groups Mean ± SD (µm2) Treatment
I 19.28 ± 2.17 x 104 PBS (0.85%)
II 6.42 ± 6.65 x 104* Pentoxifylline (ip)
III 16.94 ± 6.76 x 104 Pentoxifylline (iv)
IV 14.68 ± 2.69 x 104 Gadolinium chloride (preventive)
claimed to block phagocytosis by Kupffer cells (12-15).
In the present investigation, gadolinium initially yielded clear-cut antifibrotic results, but did not avoid the progress of fibrosis thereafter. The drug causes phenotypic alter-ation in Kupffer cells, causing them to pro-duce excess TNF-a (14). Surprisingly, our results with the India ink test did not indicate that gadolinium altered the phagocytic ca-pacity of Kupffer cells.
Vitamin A has yielded controversial anti-fibrotic results. High doses are considered to induce fibrosis in some studies (16), while others have found the opposite (17). High doses of vitamin A cause the Ito cells to be loaded with fat droplets. This may even de-press the synthetic machinery of the cell through a space-occupying effect. Our re-sults failed to show an antifibrotic effect when high doses of vitamin A were adminis-tered to rats with C. hepatica-associated sep-tal fibrosis of the liver.
Further studies with different doses and
protocols, with the present drugs or other compounds, should better characterize the antifibrotic properties of the agents used. The effects observed can also be compared in different models, since the antifibrotic activity may vary according to the pathogen-esis of hepatic fibrosis. Hepatic fibrosis is usually associated with chronic inflamma-tion and/or hepatocellular necrosis. Septal fibrosis of the liver induced in rats either by porcine serum (2,5) or by C. hepatica infec-tion (3) is not preceded by outstanding ne-crosis or inflammation. Septal fibrosis seems to have an immunological basis (18), which turns the present model still more interest-ing. In conclusion, the results of the present investigation indicate that the model of patic septal fibrosis associated with C. he-patica infection of the rat is adequate for studies concerning the response of fibrosis to drug treatment and that it may also be valuable for investigations concerning the pathogenesis of liver fibrosis.
References
1. Tsukamoto H, Matsuoka M & French SW (1990). Experimental models of hepatic fibrosis: A review. Seminars in Liver
Dis-ease, 10: 56-65.
2. Andrade ZA & Grimaud JA (1986). Evolu-tion of the schistosomal hepatic lesions in mice after curative chemotherapy.
Ameri-can Journal of Pathology, 124: 59-65.
3. Ferreira LA & Andrade ZA (1993).
Capil-laria hepatica: a cause of septal fibrosis of
the liver. Memórias do Instituto Oswaldo
Cruz, 88: 441-447.
4. Andrade ZA (1991). Contribution to the study of septal fibrosis of the liver.
Inter-national Journal of Experimental Patholo-gy, 72: 553-562.
5. Paronetto F & Popper H (1966). Chronic liver injury induced by immunologic reac-tions. Cirrhosis following immunization with heterologous sera. American Journal
of Pathology, 40: 1087-1101.
6. Junqueira LCU, Bignolas G & Brentani R (1979). Picrosirius staining plus polariza-tion microscopy, a specific method for collagen detection in tissue sections.
His-tochemical Journal, 11: 447-455.
7. Bergman I & Loxley R (1963). Improved and simplified methods for the spectro-photometric determination of hydroxypro-line. Analytical Chemistry, 35: 1961-1965. 8. Duncan MR, Hasan A & Berman B (1995). Pentoxifylline, pentifylline, and interferons decrease type I and III procollagen mRNA levels in dermal fibroblasts: evidence for mediation by nuclear factor 1 down-regu-lation. Journal of Investigative
Dermatol-ogy, 104: 282-286.
9. Kozaki K, Egawa H, Bermudez L, Keefe EB, So SK & Esquivel CO (1995). Effects of pentoxifylline pretreatment on Kupffer cells in rat liver transplantation.
Hepatolo-gy, 21: 1079-1082.
10. Romanelli RG, Caligiuri A, Carloni V, DeFranco R, Montalto P, Ceni E, Casini A, Gentilini P & Pinzani M (1997). Effect of pentoxifylline on the degradation of procollagen type I produced by human hepatic stellate cells in response to trans-forming growth factor-beta 1. British
Jour-nal of Pharmacology, 122: 1047-1054.
11. Carloni V, Romanelli RG, Pinzani M, Laffi G & Gentilini P (1996). Expression and function of integrin receptors for collagen and laminin in cultured human hepatic stellate cells. Gastroenterology, 110: 1127-1136.
12. Kamei T, Callery MP & Flye MW (1990). Kupffer cell blockade prevents induction of portal venous tolerance in rat cardiac allograft transplantation. Journal of
Surgi-cal Research, 48: 393-396.
13. Vidal C, Gonzalez-Quintela A & Cuervas-Mons V (1993). Influence of Kupffer cell phagocytosis blockade on the production of ovalbumin-specific IgE and IgG1 anti-bodies in an experimental model. Clinical
and Experimental Allergy, 23: 15-20.
14. Aril S, Monden K, Adachi Y, Zhang W, Higashitsuji H, Furutani M, Fujita S, Nakamura T & Imamura M (1994). Patho-genic role of Kupffer cell activation in the reperfusion injury of cold-preserved liver.
Transplantation, 58: 1072-1077.
15. Rai RM, Zhang JX, Clemens MG & Diehl AM (1996). Gadolinium chloride alters the
acinar distribution of phagocytosis and balance between pro- and anti-inflamma-tory cytokines. Shock, 6: 243-247. 16. Bioulac-Sage P, Quinton A, Saric J,
Grimaud JÁ, Mourey MS & Balabaud C (1998). Chance discovery of hepatic
fibro-sis in patient with asymptomatic hypervi-taminosis A. Archives of Pathology and
Laboratory Medicine, 112: 505-509.
17. Yamane M, Tanaka Y, Marumo F & Sato C (1993). Role of hepatic vitamin A and lipo-cyte distribution in experimental hepatic
fibrosis. Liver, 13: 282-287.
18. Bhunchet E, Eishi Y & Wake K (1996). Contribution of immune response to the hepatic fibrosis induced by porcine se-rum. Hepatology, 23: 811-817.
